Disney Enterprises, Inc. et al v. Hotfile Corp. et al

Filing 87

NOTICE by Columbia Pictures Industries, Inc., Disney Enterprises, Inc., Twentieth Century Fox Film Corporation, Universal City Studios Productions LLLP, Warner Bros. Entertainment Inc. re 85 Reply to Response to Motion,, PLAINTIFFS' NOTICE OF FILING REPLY DECLARATION OF DUANE C. POZZA IN SUPPORT OF MOTION TO COMPEL RESPONSES TO REQUESTS FOR PRODUCTION OF DOCUMENTS AND INTERROGATORIES (Attachments: # 1 Affidavit REPLY DECLARATION OF DUANE C. POZZA IN SUPPORT OF MOTION TO COMPEL, # 2 Exhibit A, # 3 Exhibit B, # 4 Exhibit C, # 5 Exhibit D, # 6 Exhibit E, # 7 Exhibit F, # 8 Exhibit G, # 9 Exhibit H, # 10 Exhibit I (public redacted version))(Stetson, Karen)

EXHIBIT A

From:
Sent:
To:
Cc:
Subject:

Fabrizio, Steven B
Wednesday, June 08, 2011 9:51 PM
[e-mail]
Pozza, Duane; Janet Munn; Platzer, Luke C; Andrew Leibnitz
Re: Disney v. Hotfile

Rod – We've had a chance to consider your proposal (we based our analysis on the attached file and did not
compare that word for word with what is in your email, but presume they match up).
We can live with your Point I as is.
We can live with your Point II as is except as follows: We need both the User ID information and the number
of downloads information in Step 2. Those are both necessary to identify the content files that will be in suit.
We also would tweak your "good faith" clause. We have no problem with the clause per se, just its wording.
We can only make the good faith representation as to our ownership of the copyright in the "work" that
appears to be represented by the file, not as to the "file" itself. I think that is probably what you meant to say
anyway.
The big issue — which we cannot accept — concerns the stats study data, your Point III. You incorrectly
presume that the stats study is not relevant to the DMCA defense. But the stats analyses are just as relevant
to the DMCA defense as they are to the underlying copyright issues, and serve almost the identical purposes.
First, the stats study is directly relevant to the "financial benefit" prong of the DMCA contained in 17 U.S.C. §
512(c)(1)(B). Courts look to whether there is a high proportion of infringement as evidence that infringement
is a "draw" to attract users. Almost every court beginning with Napster has relied on statistical evidence for
the financial benefit analysis. Second, as courts have held, the statistical data of the proportion of
infringement is probative of knowledge, which is obviously an important element in the DMCA analysis. Third,
the statistical evidence is relevant to the question of whether defendants have induced infringement; as two
courts (Fung and Usenet) have concluded, if defendants are found to have induced copyright infringement
under Grokster, then they are not eligible for any DMCA safe harbor. While we recognize that defendants will
argue that YouTube came out the other way on inducement and the DMCA, based on the conflicting
precedents, our Court will need to make its own decision, and, respectfully, we believe that we have the
stronger argument by far. Defendants cannot by fiat presume the Court will rule their way on a central
disputed legal issue and deny any discovery inconsistent with their legal argument (what you are suggesting
would result in piecemeal DMCA motions if our Court agrees with Fung and Usenet on the inducement
point). Accordingly, the stats analyses are directly relevant to the DMCA in that regard as well.
Defendants simply have no legal basis to refuse to produce the stats data we requested. It appears as if
defendants objective is to impede plaintiffs' ability to conduct a stats analysis because they want to suppress
that evidence. In point of fact, the stats analyses do not require substantially different data — in type or
volume — than the data that defendants are agreeing to produce in your proposal. The following is the only
marginal additional data that would be necessary:
1. For all files: the User ID and number of downloads data. Even for all files those two fields would be a

trivial amount of data.
2. For each Stats Analysis File — which would be a relative tiny subset of files, likely less than 3000

content files total: (i) a copy of the actual content file, and (ii) the IP address of the uploading user and
1

each downloading user (with the last octet masked per the ESI plan), (iii) the date and time the file was
uploaded, and (iv) the file's status per the definition in ¶ 17m of plaintiffs' Requests.
Defendants have agreed to produce all of the same categories of data (and more) for the Infringement
Analysis Files, which will be a much larger set of files than the Stats Analysis Files.
We frankly do not see how defendants can credibly justify denying us this limited data given the obvious
relevance of the stats analyses to defendants' DMCA defense.
We would like to resolve this but cannot agree to defer the stats analyses. We would ask you to reconsider
and see whether we can have a resolution based on the limited additional data we've outlined above.
SBF
From: "Roderick M. Thompson" <[e-mail]>
Date: Wed, 8 Jun 2011 13:33:17 -0500
To: Steven Fabrizio <[e-mail]>
Cc: Duane Pozza <[e-mail]>, Janet Munn <[e-mail]>, Luke Platzer <[e-mail]>,
Andrew Leibnitz <[e-mail]>
Subject: RE: Disney v. Hotfile

Steve, we have reviewed and re-ordered your proposed data production by Hotfile. We tried to accept as much of the
wording of your proposal as possible and I think you will see that at 90% of it is here (Attached is a word document showing
changes from your email--though I switched the order of paragraphs around first before marking the changes.) Please review
and get back to us promptly.
I.

Plaintiffs' Requested "DMCA Analysis" Data Production -

"DMCA Analysis File" means each of the following files: files for which Hotfile has received a takedown notice on behalf of
any copyright holder ("Takedown Notice Files"); files for which Hotfile has received an instruction to remove a file through
any Special Rights Holder Account ("Rights Holder Account Files"); files that are a component of the same multi-part archive
(e.g., RAR) of a Takedown Notice File or a Rights Holder Account File; files that are copies of a Takedown Notice File or Rights
Holder Account File made using the Hotfile multiple copy feature; and files otherwise disabled for copyright infringement.
For each "DMCA Analysis File" (defined below), Hotfile would produce the following: (i) the file name, (ii) the unique Hotfile
identifier, (iii) the Hotfile URL, (iv) the IP address of the uploading user (with the last octet masked per the ESI plan), (vi) the
Hotfile User ID for the uploading user, (vii) the status of the user (e.g., active, terminated, suspended) and, for any status
other than fully active, data indicating the date of and reason for each status change, (viii) the file's status per the definition
in 17m of plaintiffs' Requests, and (ix) as applicable, the date the file was rendered inactive. This is the minimum data
necessary for plaintiffs to analyze whether defendants have complied with DMCA notices and the DMCA's threshold
requirement that defendants reasonably implement a repeat infringer policy, 17 U.S.C. § 512(c)(i)(1)(A).
II.

Plaintiffs' Requested "Infringement Analysis Files" Data Production -

Step 1: For all files on Hotfile, regardless of the file's status (e.g., active, inactive), defendants would produce the following
data: (i) file name, and (ii) unique Hotfile identifier.

2

Through analysis of this data, plaintiffs will identify "files that appear to contain plaintiffs' copyrighted works ("Infringement
Analysis Files")".
Step 2: For each Infringement Analysis File, Hotfile would produce the following: (i) the Hotfile URL, (ii) the file size in bytes,
(iii) the IP address of the uploading user and each downloading user (with the last octet masked per the ESI plan), (iv) the
Hotfile username for the uploading user, (v) whether the file was a copy made using the Hotfile multiple copy feature, (vi) the
date and time the file was uploaded and of each download, and (vii) the file's status per the definition in 17m of plaintiffs'
Requests. Through analysis of this data, plaintiffs will identify for defendants a subset of files for which plaintiffs' need the
actual content file ("Selected Content Files"), for each of which Plaintiffs will represent they have a good faith belief that they
own a valid copyright.
Step 3: For each Selected Content File, Hotfile would produce a copy of the actual content file.
III. The "Stats Analysis File," would be deferred without Prejudice to both parties' positions. Hotfile believes that there is no
reason to undergo this analysis, especially unless the Studios survive an early DMCA summary judgment motion. Plaintiffs
are of course free to do whatever analysis they would like of the data produced under Category I and II; there just is no
present justification for additional data to be provided solely for "statistical purposes." (Alternatively, we are willing to
discuss a more relevant statistical analysis crafted jointly by both of our experts.)

-----Original Message----From: Fabrizio, Steven B [mailto:[e-mail]]
Sent: Friday, June 03, 2011 5:08 PM
To: Thompson, Rod (27) x4445
Cc: Pozza, Duane; [e-mail]; Platzer, Luke C; Leibnitz, Andrew (21) x4932
Subject: RE: Disney v. Hotfile

We see the events leading up to our motion quite differently but, as you say, enough said.
As for your question:
- The data regarding the number of times a file is downloaded is necessary because a file downloaded 100,000 times and a
file downloaded 3 times would not be treated the same in selecting a "representative" sample of files distributed by
Hotfile. The process is more complicated than a simple random sample. There would be weighting of files as part of the
sample selection protocol.
- The user ID is needed because of the prevalence of RAR files, especially for larger sized files. A movie, for example, might be
uploaded to Hotfile as three separate RAR files. For statistical purposes, we don't treat each RAR file separately, but rather
measure "content" files. So in my example the three RAR files constituting a single movie would count as one "file" for
purposes of the stats analysis. We do this as part of our taking every precaution to ensure that the sample is not biased in
our favor. (An infringing movie uploaded as three RAR files counts once, not three times.) The user ID is necessary to
associate the RAR files into their constituent content files. The RAR files are identified by part (part 1, part 2, part
3). Without the user ID we might associate the wrong parts together (since files might be named the same); with the user ID
we can be confident that we have associated the right RAR files to make up the content file that will be part of our study.
If I haven't explained that clearly enough, just give me a call tonight or over the weekend (my cell is 703-307-7125).
I'm glad you are giving the proposal some consideration. I need to emphasize, however, that we didn't make that as an
opening salvo from which a compromise can be reached. That is really the bottom line of what we need. If it is defendants'
intention to try to chip away at the data we've proposed, don't waste a lot of your own time. We cannot take less than what
is in that proposal.
Have a good weekend.
3

SBF

-----Original Message----From: [e-mail] [mailto:[e-mail]]
Sent: Friday, June 03, 2011 7:27 PM
To: Fabrizio, Steven B
Cc: Pozza, Duane; [e-mail]; Platzer, Luke C; [e-mail]
Subject: RE: Disney v. Hotfile
Steve, first, I saw no productive reason to rehash in my letter what happened. (Andy had explained our views to Duane, I
believe very clearly.) But since you are baffled--we were not happy with what can only be called a blatant sandbag. When
you asked us to preview the grounds for our motion to dismiss so your team could make vacation plans (we told you we dold
not challenge personal jurisdiction and therefore you did not need to worry about immediate discovery), we obliged even
though we did not have to and might have thus inconvenienced your team. But that is not the kind of "tactical" advantage
we seek and is not productive in the long run to efficient case resolution. When we asked last Friday, the day before a long
holiday weekend, if your team would be filing a motion to compel we got a non-answer. Instead the call ended with setting a
date for further meeting and conferring. That is why we were surprised and disappointed by the filing the next business
day. Enough said.
I will also resist responding to the many unnecessary accusations in your email and will not attempt to correct the recounting
of who said what when, as that would be a still further waste of time and probably prompt another long response from
you. Just remember this is not Fung or Bunnell and it really doesn't matter what happens in what you believe are
"comparable" file sharing cases. This is a legitimate file hosting service, more comparable to YouTube and Rapidshare.
Let me clear up one misperception: when I said your motion hit "shortly after I had confirmed" that hotfile would agree to
the compromise I outlined, I meant they had confirmed to me that I had the authority to offer the compromise. (We were
still under the impression that our meeting and conferring was ongoing and that it was a good use of our time to look for a
compromise.) I did not mean to imply I had confirmed it with anyone on your team; I never had a chance to.
Notwithstanding the content of the long preamble, we appreciate your detailed proposal and will give it serious
consideration. That will take some time because of the weekend, time zone differences and the complexity of what you
propose. One question--why is the number of times a file is downloaded and the uploading user ID needed to select a
random sample of files?
Enjoy your weekend.
Rod
-----Original Message----From: Fabrizio, Steven B [mailto:[e-mail]]
Sent: Thursday, June 02, 2011 8:25 PM
To: Thompson, Rod (27) x4445
Cc: Pozza, Duane; Janet Munn; Platzer, Luke C; Leibnitz, Andrew (21) x4932
Subject: Re: Disney v. Hotfile

Rod - I've had a chance to review your June 1 letter and discuss it with my team to make sure all of our recollections were
consistent. I have to tell you that I am baffled by your comment in your cover email that your team "was not happy" about
our motion. Why? We could not have been clearer (or more repetitious in saying) that, on the core issues, where we had
received essentially categorical refusals to produce by defendants, we intended - and needed - to move to compel as quickly
as possible. You certainly could not have been surprised.

4

Regrettably, your letter contains material factual inaccuracies and, in its omission of other facts, is misleading. We accept
that this was not done deliberately, but it does put an inaccurate and unfair spin on what, to us, are pretty clear-cut
facts. The facts, as we recollect them, are as follows:
First, our meet and confer on all issues we raised with the Court was completed, exhaustively so, and that fact was
acknowledged by both sides. We had exhausted discussion, negotiation and the possibilities of compromise, and had
mutually concluded that there was simply no agreement to be reached on those categories of discovery. You personally had
all but told us that your clients would never allow you to voluntarily produce those categories of data, documents and
information. In conversations in which you did not participate, Andy confirmed, unequivocally, that, on those topics,
defendants were standing on their objections and no further discussion would change that. So, while the parties may be
continuing discussions on other (wholly different) categories of discovery, on which further discussion may result in
compromises acceptable to both sides, it is inaccurate to suggest that the meet and confer on the topics that are the subject
of our motion was anything but fully completed.
Indeed, our respective teams met and conferred on those topics no less than four separate times, on May 11, May 16, May
19, and May 20. You will also recall that plaintiffs tried very hard to meet and confer even before the parties exchanged
written responses, so as not to waste time on objections that readily could be avoided. Defendants refused. The topics on
which we moved go to the heart of discovery and defendants' failure to produce this discovery is materially impeding our
case preparation. Plaintiffs have been trying to keep discovery moving apace for the obvious reason that their copyrighted
works are being infringed thousands of times over every day by defendants. On top of that, defendants have insisted on
moving for leave to file an "early DMCA motion" by July 15. In light of the pendency of that motion, how long would you
expect plaintiffs could wait to ask the Court to order discovery of key categories of information, categories which are
routinely produced in comparable cases?
Second, your account of the "idea" you "floated" does not conform to our recollection (except insofar as you made
abundantly clear that your clients had not approved it). The proposal you make in your letter is not even the same as the
proposal you suggested during our May 16 meet and confer. On May 16 you asked whether we could accept data containing
far fewer fields than we requested and need. We told you that we needed the data for multiple analyses, most notably a
statistical study and a direct infringement analysis (as we also explain in our motion). Your letter continues to misapprehend
the purposes for which we need the data and, therefore, continues with proposals that cannot work.
Nevertheless, during the May 16 meet and confer, we examined the requested fields and identified the minimum fields we
would need: the Hotfile URL (on assumption, to be confirmed, that the URL systematically included the Hotfile identifier and
the file name), the file size in bytes, the location and/or IP address of the uploading user, the number of downloads, the
location or IP address of the downloading users, and the status of the file. That was the minimum we believed we needed
just to conduct our direct infringement and stats analyses (we would need additional subsets of data for other important
analyses, e.g., a repeat infringer analysis and an analysis of compliance with takedown notices). On our May 16 call, you
immediately dismissed producing the data we identified and instead said you would check with your clients as to whether
they would produce just the Hotfile URL and file size data. As we told you, that suggestion would complicate the direct
infringement study and likely would compromise the statistical analyses. On the direct infringement front, it would require a
multiple step process in which we first identify innumerable URL links that appear to correspond to plaintiffs' copyrighted
works (a process that would likely result in our identifying tens of thousands of links). Then we would have to make a
subsequent request for defendants to produce the remaining fields of data for those identified URL links. It is worth
observing that on May 16 you would not in fact commit to producing those other fields in the second part of the process and your June 1 letter remains studiously silent on that key question. As for the statistics study, we told you that the way our
expert would identify a representative sample would require having some of the other fields of data.
That said, we said we would work with our statistics expert to consider alternative sampling protocols - but only after you
first confirmed with your clients and advised us that your clients would agree to your proposal, including committing to the
necessary follow-up production of data. Since you opened the May 16 discussion of this "idea" by saying that you doubted
your clients would agree to produce any of this data, it hardly made sense for us to work with our expert on sampling
protocols if your clients were going to reject your recommendation.
We never heard back from you. It is a little over the top for you to complain in your June 1 letter that you never heard back
from us. We told you that we did not think your proposal would work. We also told you that we could not see any reason
5

not to produce all the data fields we requested so that we could conduct our studies without potentially compromising
them. You made it abundantly clear that your clients would never go for that and probably would reject the idea of
producing even the much narrower set of data. Frankly, from the way you described the prospects of getting approval from
your clients, we didn't really expect to hear back from you. Still, our teams had three meet and confer teleconferences after
that May 16 discussion and before we filed our motion, including one on May 27 that you personally participated in for some
of the discussion. No one from your team ever mentioned the idea that defendants might agree to produce a narrower data
set.
To the contrary, on May 19, we had a meet and confer call with Andy, in which Andy made it clear that defendants were
standing on their objections as to the Content Reference and User Data. We did our best to persuade him otherwise,
reminding him that the courts in the Fung and Bunnell cases had ordered defendants to produce comparable data. Andy
asked us to send him the orders, while simultaneously telling us unequivocally that defendants' position would not
change. The next day, May 20, in an email time-stamped 10:57am, Duane forwarded the two orders, with an email that
confirmed the state of affairs:
"We also indicated that we would send copies of orders compelling production of user activity data. I am therefore attaching
orders from the Bunnell and Fung cases. The Bunnell order was upheld by the district court at 245 F.R.D. 443. As we
discussed, the parties are currently at an impasse on producing complete Content Reference, User, and Affiliate Data, and
defendants have given no indication of changing their position. We are sending these as a courtesy but do not understand
that our meet-and-confer is being reopened on this issue." (Emphasis added.)
No one from your team ever took issue with that statement because it accurately reflected what had been mutually
concluded in the meet and confer. Certainly no one ever mentioned your "idea" from the May 16 meet and confer.
Third, your June 1 letter implies that you confirmed to us the day before we filed our motion that your client had agreed to
produce a narrower set of data. That of course is not the case. The first we heard of that was on June 1 - the day after we
filed our motion - when Andy so advised Duane. Even if your clients had agreed to produce the full data set that plaintiffs
need, agreeing to produce discovery after forcing the other side to file a motion does not excuse the non-production. Fed. R.
Civ. P. 37(a)(5)(A). But, here, defendants have not offered to produce the data set we requested and need. You are
suggesting producing just the "file name and unique identification number" for Hotfile-hosted content, which is even less
than you suggested in the May 16 meet and confer. As we told you more than two weeks ago, that framework would not
allow us to conduct our statistical analysis. It also creates a needlessly cumbersome multiple step discovery
process. Defendants notably also have not committed to the follow-on discovery that we discussed would be
necessary. Bottom line: Your proposal, as stated in your June 1 letter, simply does not move us closer to a compromise that
plaintiffs can accept, as we suspect you already knew.
Maybe we have become jaded but we cannot help but think that defendants have made an offer they knew would be
unacceptable for whatever tactical value defendants believe it might have in connection with our pending motion to
compel. Nevertheless, genuinely hoping we are wrong, we've developed a proposal that could work. It is far more
complicated than simply producing the full data sets that plaintiffs have requested. And there is no reason for this more
complicated, multi-step approach because, as a practical matter, producing the full data sets would almost certainly be far
less burdensome for both sides. Still, subject to consultation with our clients and statistics expert, which we will do if and
when you advise us that defendants would produce discovery according to this plan, plaintiffs would agree to the following
plan - involving data production in three steps - to resolve all outstanding issues with regard to our Request Nos. 1 (Content
Reference Data) and 2 (User Data).
HOTFILE DATA PRODUCTION - STEP 1
* For all files on Hotfile, regardless of the file's status (e.g., active, inactive), defendants would produce the following
data: (i) file name, (ii) unique Hotfile identifier, (iii) the Hotfile user ID for the uploading user, and (iv) the total number of
times each file has been downloaded. (The data regarding uploading user ID and number of downloads is necessary for
plaintiffs to conduct their statistical analyses.)

6

* Through analysis of this data, plaintiffs will identify for defendants two sets of files: files that appear to contain plaintiffs'
copyrighted works ("Infringement Analysis Files") and files that may be used in plaintiffs' statistical analyses ("Stats Analysis
Files").
HOTFILE DATA PRODUCTION - STEP 2
* For each Infringement Analysis File, Hotfile would produce the following: (i) the Hotfile URL, (ii) the file size in bytes, (iii)
the IP address of the uploading user and each downloading user (with the last octet masked per the ESI plan), (iv) the Hotfile
username for the uploading user, (v) whether the file was a copy made using the Hotfile multiple copy feature, (vi) the date
and time the file was uploaded and of each download, and (vii) the file's status per the definition in ¶ 17 of plaintiffs'
Requests. Through analysis of this data, plaintiffs will identify for defendants a subset of files for which plaintiffs' need the
actual content file ("Selected Content Files").
* For each Stats Analysis File, Hotfile would produce the following: (i) a copy of the actual content file, and (ii) the IP address
of the uploading user and each downloading user (with the last octet masked per the ESI plan), (iii) the date and time the file
was uploaded, and (iv) the file's status per the definition in ¶ 17 of plaintiffs' Requests.
* In addition, for each "DMCA Analysis File" (defined below), Hotfile would produce the following: (i) the file name, (ii) the
unique Hotfile identifier, (iii) the Hotfile URL, (iv) the file size in bytes, (v) the IP address of the uploading user and each
downloading user (with the last octet masked per the ESI plan), (vi) the Hotfile username for the uploading user, (vii) the
status of the user (e.g., active, terminated, suspended) and, for any status other than fully active, data indicating the date of
and reason for each status change, (viii) the file's status per the definition in ¶ 17 of plaintiffs' Requests, and (ix) as
applicable, the date the file was rendered inactive. This is the minimum data necessary for plaintiffs to analyze whether
defendants have complied with DMCA notices and the DMCA's threshold requirement that defendants reasonably
implement a repeat infringer policy, 17 U.S.C. § 512(c)(i)(1)(A).
A "DMCA Analysis File" means each of the following files: files for which Hotfile has received a takedown notice on behalf of
any copyright holder ("Takedown Notice Files"); files for which Hotfile has received an instruction to remove a file through
any Special Rights Holder Account ("Rights Holder Account Files"); files that are a component of the same multi-part archive
(e.g., RAR) of a Takedown Notice File or a Rights Holder Account File; files that are copies of a Takedown Notice File or Rights
Holder Account File made using the Hotfile multiple copy feature; and files otherwise disabled for copyright infringement.
HOTFILE DATA PRODUCTION - STEP 3
* For each Selected Content File, Hotfile would produce a copy of the actual content file.
We cannot emphasize enough that this is a far more complex and burdensome process - for both sides - than what plaintiffs
had requested or, frankly, would prefer. Nevertheless, if defendants will stipulate to produce the data identified in each step
on an expeditious basis, so that plaintiffs are not unduly prejudiced by the multi-step approach, then this could be a basis for
plaintiffs to withdraw their motion to compel as to RFP Nos. 1 and 2, i.e., Section I of our motion to compel.
Let us know if defendants are prepared to so stipulate.
SBF
From: "Roderick M. Thompson" <[e-mail]<mailto:[e-mail]>>
Date: Wed, 1 Jun 2011 22:36:59 -0500
To: Steven Fabrizio <[e-mail]<mailto:[e-mail]>>
Cc: Duane Pozza <[e-mail]<mailto:[e-mail]>>, Janet Munn
<[e-mail]<mailto:[e-mail]>>, Luke Platzer
<[e-mail]<mailto:[e-mail]>>
Subject: Disney v. Hotfile

7

Steve, please see the attached and call me if you'd like to discuss. Our team was not happy about this . . .
Rod
Roderick M. Thompson
Attorney at Law
______________________________
Farella Braun + Martel LLP
RUSS BUILDING
235 MONTGOMERY STREET
SAN FRANCISCO / CA 94104
______________________________
T 415.954.4400
D 415.954.4445
F 415.954.4480
www.fbm.com

________________________________
Steven B. Fabrizio
Jenner & Block LLP
1099 New York Avenue, N.W.
Suite 900
Washington, DC 20001-4412
Tel [phone]
Fax [phone]
[e-mail]
www.jenner.com<http://www.jenner.com/>

CONFIDENTIALITY WARNING: This email may contain privileged or confidential information and is for the sole use of the
intended recipient(s). Any unauthorized use or disclosure of this communication is prohibited. If you believe that you have
received this email in error, please notify the sender immediately and delete it from your system.

________________________________

8